Sound Practices Magazine Online!

Audio Note Ongaku
Article by Hiroyasu Kondo-san, Audio Note
From Sound Practices Issue 2, Fall 1992

  The 211 creates a formidable impression because of its shape. But the 211 may appear old-fashioned and unsophisticated to high-brow maniacs for audio equipment. Perhaps this is the reason why there are only a few amplifiers that employ the 211 tube today.

This type of amplifier is a challenge from the viewpoint of the manufacturer. The power supply circuit cannot be normally designed, since the supply voltage required for the 211 tube must be approximately 1000V. In manufacturing the amplifier, we must pay utmost attention to moisture-proofing the chassis and proper insulation of the wiring so as to prevent hands from getting electric shock if one reaches into the amplifier.

However, it is only large triodes like the 211 that provide the best performance that vacuum tubes can offer. The Eg-Ip characteristic of the 211 and its brothers is more linear than most other power tubes. Larger diameter wire is used for the heater, the grid to heater spacing is considerable, and coarse "pitch" is used for the grid winding geometry. This type of vacuum tube is very sturdy and promises a high vacuum integrity, since it was originally intended for demanding commercial and military applications.

When high plate voltage is employed and the pitch of the grid is made more coarse, the amount of "stray electrons" around the heater is greatly decreased, thereby allowing the vacuum tube to work in a more closely ideal manner. In particular, the large triode vacuum tube functions mare perfectly when dealing with low level signals, allowing the tone quality to be clear and distinct. If the design of the driver stage is satisfactory, the single triode power amplifier possesses an unbelievably law distortion characteristic, even in the absence of negative feedback. This fact demonstrates the excellent performance of this family of vacuum tubes.

Simple Is The Best
What does 'good sound' mean? There may be a variety of interpretations. However, a good sound must mean a "natural sound". The meaning of "natural sound" depends on the interpretation of the word. There may be differing opinions -- such as "natural" means a sound "without mechanical sound" or that it means a sound "alike to live sound". All listeners would agree that natural sound must not include mechanical sound. A mechanical sound will paralyze the listener's senses when listening to the music for an extended period of time. There are many factors which can produce mechanical sound 

It is an interesting discovery that when mechanical sounds are thoroughly eliminated, we can reproduce live sound. What must we do to eliminate mechanical sound? First, we must simplify the audio circuitry. Next, we must carefully select the components which are used in the equipment.

At the closing stage of the vacuum tube amplifier age, tube circuit technology was in a highly advanced state. It was like the advanced evolution of dinosaurs at the close of the dinosaurian age. When I was only 20 years old, I concentrated my energies on the study of "negative feedback theory". At such an inexperienced age, one tends to concentrate on a meeting point, where theory seems to meet reality, instead of focusing on the sound itself. I still remember that I was trying to make up an ideal circuit so as to produce a good sound and my soldering iron was always in its heated state. In spite of my efforts, I could not go beyond the wall of "good" sounds. I finally arrived at the conclusion, after considerable trial and error, that the most important thing is to "make up an amplifier without depending on negative feedback." In fact, I found that I had to strive to simplify the circuitry even without negative feedback and also to develop my understanding of the characteristics of materials to be used and their effect on sound quality.

At one Audio Fair last year, we compared sound reproduced by a transistor amplifier produced by a famous manufacturer and sound reproduced by a 211 vacuum tube amplifier of apparent good quality. To my surprise, the 211 amplifier sounded exactly like the transistor amplifier! I expected more from the triode. Perhaps this was due to the use of feedback in the triode amplifier. Clearly to raise the performance of a single triode amplifier above the crowd it is necessary to optimize the design and parts quality used in its realization, It is not enough simply to use triodes.

I reemphasize that it is necessary to simplify the fabrication of an amplifier and to carefully select highest quality parts in order to reproduce a "quality of live sound" free from mechanical sound. These are the most important considerations in producing an ideal amplifier.

I Am An Enthusiast For Single-Ended Amplifiers
 The single type power amplifier enables greater simplicity compared with other types of amplifiers, if high power is not a requirement. Even a non-feedback single ended amp can provide low distortion with proper selection of power tubes. With the right parts, the sonic results can be excellent and we can enjoy a more dynamic sound than that produced by a higher powered amplifier. If the circuit is simplified, in other words, if the number of parts to be used is reduced even by one, the sound produced by that part, mechanical sound, will be eliminated.

It is interesting that we cannot simply say "the single type can reproduce a good sound." Some single type amplifiers can produce more distortion. Choice of tube types is important in this respect. Distortion in the power tube cannot be eliminated completely, even if the previous stage is designed for minimum distortion. Push-pull circuits can eliminate the second harmonic produced in the output stage. This is the reason why many high class amplifiers employ the P-P configuration.

However, some say that a push-pull output stage makes the sound unclear. There is certainly a potential for unclear sound in these designs due to imbalances. The current state of parts, especially the OPT, is not sufficiently advanced to avoid this problem. The deterioration of sound quality due to improper balance is a challenge for the designer to solve 30 years from now. On this basis, the single type can be recommended, since this type is free from unclear sound due to imbalance.

However, there is a specific challenge which the single type must overcome and we cannot say that the single type is good unless it successfully overcomes this problem. In short, we cannot blindly become enthusiasts for single ended amplifiers.

Problem With Output Transformers
Through the single type OPT, much direct current passes. If an iron core is fixed to the coil and DC passed through the coil, the iron core will be magnetized as a result, thereby causing deterioration of the characteristics of the low end of the sound range. Magnetization may lead to low-pitched sounds not being produced at all. In order to avoid the hazardous magnetization of the core by DC, an air gap is provided in the magnetic path of the iron core. In order to enhance bass reproduction, the volume of the iron core must be increased so as to increase the magnetic flux handling capability of the transformer.

The world is as kind as it is cruel. The loss generated by the air gap brings the magnetic characteristics to a crisis (Note the hysteresis curve), thereby causing the smooth curve to be changed into a curve which is partially linear and gradually less and less linear as it approaches saturation. Actually, this action enhances the linearity of the mu (permeability) in the small signal region. The enhanced linearity in the low level signal range improves the transmission capability of small signals, thereby enabling greater clarity and detail in musical reproduction. This phenomenon does not occur in push-pull transformers.

Merely increasing the volume of the iron core, i.e. increasing the size of the sectional area of the iron core, will not necessarily lead to improvements in musical reproduction. There remains the problem of the tone quality reproduced by the material of the iron core. This is the reason why our company manufactures the OPTs at our own plant. Needless to say, we carefully select the material to be used for the iron core (by placing an importance on the resolution of sound) so as to reproduce good tone quality. Our transformer employs a cut core. A non-magnetizing band is used for the core as well as a non-magnetic transformer case. We pot our transformers with 24 hours-curing type epoxy to fix the coil and core securely to avoid mechanical resonances. The wire is wound around the bobbin by hand so as to provide a proper tension in the wire. No other OPT in the world is manufactured with such attention to the resulting quality of tone.

Matching Of The OPT And Power Tube
The capability of an OPT is typically indicated in terms of low end performance, for instance, wattage that can be transmitted at 50 Hz with 3% distortion. The type of distortion introduced in the OPT is distortion of the AC signal current, a most troubling type of distortion. The level of distortion produced in the output stage varies with the relation of the plate resistance of the power tube to the load impedance reflected to the primary of the transformer. The OPT for the Ongaku provides excellent performance through the entire audio range, because the primary impedance is a very high (20 kOhm) and first class silver wire is used. The size of this OPT is large beyond common sense.

This single triode amplifier displays outstanding low end characteristics when compared with other amplifiers of this type because the OPT core volume is very large. It can successfully reproduce a very soft and warm low-pitched sound. We tend to imagine that we can enhance the tone quality if we were to utilize high permeability core material. But permalloy has a weak point in that it cannot provide enough power in low notes, since it will saturate at a magnetic flux density below the operating point of power amplifiers. In order to reproduce good quality sound, we must carefully select the iron core material, carefully specify the winding method, and choose the material of the transformer wire with optimal sonic results in mind.

Condensers
A coupling condenser is often necessary in vacuum tube amplifier circuits. The measurement of the internal series resistance of the condenser often shows an unexpectedly high value. This is not an exceptional case in a small capacity condenser. The cause of this problem is in the structure of the condenser.

The largest problem with the tone quality of the condenser is a "noise between electrodes" - audible energy created by the above-mentioned resistance. Once this noise is produced, the electrode-to-electrode area will produce noise continuously. It adversely affects the tone quality because of its high Q resonance. The reproduced sound will become a metallic sound or a frivolous sound just like a rubbing sound produced by high-polymer films. The problem is part electrical and part mechanical.

The problem with condensers is not limited to crude forms of distortion. We are manufacturing condensers with particular emphasis on materials in order to minimize sonic deterioration. Specifically we use a thick silver-leaf for the electrode and wind the foil by hand making sure to provide enough tension. I still remember my deep emotion when I first listened to an amplifier using the silver-leaf condenser. The tone quality of this vacuum tube amplifier is so excellent that it is unparalleled in the world. The reproduced sound is, needless to say, free from metallic sound. A very dynamic sound effect and sound approaching live sound can be reproduced. The pizzicato of low strings can be reproduced properly in time. The reproduced sound really offers "music." Although the higher cost of the silver-leaf condenser is a disadvantage, the use of this type condenser is inevitable when the goal is to reproduce a really good sound. This must be a ‘favorite amusement of a king."

Circuit of The Ongaku Amplifier
The circuit itself is unexceptional. The 211 is driven by SRPP input stage, a common type voltage amplifier comprises the second stage and a cathode follower provides the third stage. The first stage is DC coupled to the second stage, as are the driver tube and output stages. A high supply voltage around 500V is applied to the second stage so as to ensure enough driving voltage for the 211 grid. On the third stage, the minus power supply supplies the voltage to the cathode so as to regulate bias voltage. An important feature of this circuit is its power supply circuit. A full wave bridge is constructed from four 5AR4 tubes with sections parallel to rectify the B+ supply. We use the utmost care when selecting the vacuum tube, since the performance of the vacuum tube determines the quality of sound to be reproduced.

Silver Wire Wound Output Transformer
The most significant mistake of theoretical audio maniacs in this century is that they threw away the OPT. In many cases, transformers can adversely affect the quality of reproduced sound but I feel it is not correct to employ the OTL without conducting adequate experimentation and without particular attention to design considerations.

Although they say that the resolution may be degraded if a signal is transmitted through the OPT, it is a well-known fact that the transformer can offer a good feeling of the music that the OTL cannot. Accordingly, we are required to solve the problem of how to enhance the resolution of sound with OPT.

I can still remember my deep emotion when I reproduced the sound using the MC step-up transformer wound with silver wire 13 years ago. Using silver, I could successfully resolve the "problem with the noise produced by the transformer," which we could not solve up to that time in the recording studio. From that time, we began using silver wire in a wider range of applications. We finally arrived at the conclusion that magnetism matches the silver wire better than the copper wire. The performance of the silver wire used in the magnetic circuit is wonderful. The performance gives us a feeling as if a signal which once got blocked rushes forward like floodwater. We can perceive that the signal flows smoothly without binding. The work of the silver wire is especially wonderful in the small signal area. The silver wire wound transformer contributes to the enhancement of tone quality reproduced by the vacuum tube amplifier, thereby allowing the vacuum tube amplifier to become famous for its good quality of reproduced sound. In some of our amplifiers, such as the KEGON/C, copper wire is used for the primary side of the OPT to reduce cost and only the secondary is wound with silver wire.

Others
The chassis of the Ongaku amplifier is made of pure copper for the purpose of enhancing the tone quality. The twin-monaural configuration is employed for the purpose of minimizing the interference between the sounds reproduced from the left side and the right side. The realization of such purposes resulted in a larger and heavier chassis. One of the reasons why the "Ongaku" amplifier won world wide fame is the elaborate execution of this amplifier.

Epilogue
As a result of our endeavor, the completed 211 can really reproduce "musical" sounds. It is an optimal amplifier for listening to classical music. It not only reproduces a live sound but it also allows the listeners to identify what part of the music is reproduced from what position of the equipment. For instance, when we reproduce a violin concerto, the melody played by the violin is reproduced more prominently than other sounds. In addition, the reproduced music is free from any metallic sound. We can listen to the bright vocal as if enjoying a live concert. I never saw such a highly sophisticated power amplifier before. The 211 "Ongaku" amplifier informs the listener with new life. How can we express such an excellent and majestic sound as this reproduced sound? We can only say "Ongaku!" to describe such reproduced sounds.

We must confess that it is not an easy task to master the secrets of the audio equipment!

Ongaku For The Rest Of Us
I don't know about you, but I sure can do without that "mechanical sound" characteristic of complicated designs. Unfortunately, an authentic Audio Note Ongaku is not In this year's budget. Actually, the retail price of this amp may exceed my total net worth by a slight margin -- even exclusive of shipping charges. If I sell my present system, I might be able to swing a pair of those hand-rolled silver coupling caps though.

But wait... because Kondo-san was gracious enough to share his design with the international experimenter community. we can build our own 211 SE amps based on the Kondo masterpiece. Of course, we can't build a real Ongaku -- part of the celebrated sonic magic of the amp is in its exquisite material execution and the special virtues of in-house Audio Note parts. But I'll bet the Kondo circuit will turn in a respectable performance even when more pedestrian materials are employed. Until the Sound Practices publishing empire hits the "big time", I'm afraid I must settle for second best when a savings of $60,000 is involved.

Some interesting components are available from Audio Note, UK which would be appropriate for a "Poor Man's Ongaku" project. including single ended transformers for the 211 and a line of oil/paper coupling caps. A primary mission of the Audio Note UK operation is to make some of the findings, of Kondo's uncompromising explorations available to the audiophile who isn't blessed with an Ongaku-sized budget. Toward this goal, they are developing a line of lesser-priced equipment of high musical achievement along with providing a selection of music quality parts for fellow audio adventurers.

OK, homebrewers... time to get those soldering irons "in the heated state", as Kondo-san puts it. Next time: my report on "Building the Poor Man's Ongaku" and more on this interesting design.

Click here to view the Ongaku schematic.